What Is Schema Markup?

How schema markup works

Schema.org is a shared vocabulary launched in 2011 by Google, Microsoft, and Yahoo, with Yandex joining later that year. Today it is developed through an open community process, with the founding search engines still acting as sponsors. The vocabulary defines hundreds of types, such as Organization, LocalBusiness, Product, Event, and Article, plus the properties each type can have, such as name, address, telephone, or openingHours.

To add schema markup, you pick the type that best matches the page, fill in its properties with facts that appear on that page, and place the result in the page’s code. There are three common formats. Microdata and RDFa add labels directly to the page’s HTML (the code that structures a web page). JSON-LD, the format Google recommends, keeps everything in one separate block of code, which usually makes it the easiest to maintain. Many website platforms and SEO plugins generate basic markup automatically, which is a reasonable start but often generic.

Schema markup is one part of technical SEO. It supports clear, well-organized page content rather than replacing it. Some AI search tools may read it too, but how they use it varies by product and is not fully documented.

What a typical business site marks up

Most business sites need only a handful of types, applied where they genuinely fit:

• Organization on the homepage or about page: name, logo, website, contact details, and sameAs links to official profiles.
• LocalBusiness, or a more specific subtype, on each location page: address, phone, hours, and map coordinates.
• Product with an Offer on product pages: name, image, price, and availability.
• Event for workshops, webinars, or open houses with a set date and location.
• Article or BlogPosting on editorial content: headline, author, and publish date.
• BreadcrumbList on pages that show a breadcrumb trail.

The sameAs property is worth extra attention. Pointing it at your official profiles and listings can help machines recognize that your site and those pages describe the same business, a core idea in entity SEO. Accuracy matters more than volume: a few correct, consistent blocks are worth more than dozens of half-filled ones.

Common mistakes to avoid

• Picking a type that doesn’t fit. Marking up a category page full of products as a single Product, or settling for a generic type when a closer one exists, misdescribes the page. When no single type fits, list more than one, as in the clinic example.
• Self-serving review markup. Google treats reviews of a business placed on that same business’s website, whether marked up directly or pulled in through a third-party widget, as self-serving and does not show review stars for them, so marking up your own testimonials earns nothing in Google Search.
• Broken or incomplete code. One missing comma or bracket can make a JSON-LD block unreadable, and a block that reads correctly can still lack a property a search feature requires. Test after every change.
• Conflicting blocks. A plugin and a hand-added block can describe the same business with different names, hours, or phone numbers. Pick one source for each fact.

The broader rules, such as marking up only what visitors can see and updating markup when the page changes, apply to all structured data.

Schema markup vs. structured data

Structured data is the broader idea: information organized in a standard format so machines can read it reliably. Schema markup is one specific and very common way of adding structured data to web pages, using the Schema.org vocabulary in a format such as JSON-LD. People often use the two terms interchangeably, which is usually harmless in everyday SEO conversations, but structured data also includes formats that have nothing to do with Schema.org, such as microformats or Open Graph tags on web pages and, in the broader sense, spreadsheets, databases, and product data feeds. In short, all schema markup is structured data, but not all structured data is schema markup.